Identification and Fault Diagnosis of nonlinear dynamic processes using hybrid models

This work addresses a novel approach for fault diagnosis of industrial processes using hybrid models. A nonlinear dynamic process can, in fact, be described as a composition of different affine sub-models selected according to the process operating conditions. This paper deals with the identification of hybrid model parameters through input-output data affected by additive noise. The fault detection scheme adopted to generate residuals uses the estimated hybrid model. In order to show the effectiveness of the developed technique, the results obtained in the fault diagnosis of a real industrial plant are reported.